Presidential campaigns are not usually about poverty. It’s a neglected issue, not least because most voters, even in Democratic primaries, live above the poverty threshold, and so economic policy tends to be discussed in terms of helping the “middle class” or “American families,” not the poorest of the poor.
2020 is shaping up to be a little different. Several Democratic presidential candidates have coalesced on a distinctive approach to fighting poverty — transferring people money — and are including it as part of bigger proposals that help poor and middle-class people alike, ensuring that these ideas have a more prominent spot in the campaign than past anti-poverty plans have had.
So far, rumored and announced candidates have outlined at least five formal proposals designed to direct additional cash to low-income households. These proposals, all co-authored by senators — Cory Booker, Kamala Harris, Sherrod Brown, and Michael Bennet (Rep. Ro Khanna is a co-sponsor on one) — attempt to do different things: Two would expand the earned income tax credit (EITC), another two offers assistance for rent, and one is a child allowance. But they all have the same overriding objective: improving the lives of low-income Americans.
The press tends to analyze proposals like these in terms of how much they cost, how realistic their passage is, and so on. And while that’s important, I had a different question: Which of these plans would do the most to cut poverty in America?
It turns out that a team of researchers at Columbia University’s Center on Poverty and Social Policy was asking that question too, and they have the skills to actually answer it. They estimated how much each of these five bills would reduce poverty, deep poverty (measured as the share of people living at or below 50 percent of the poverty line), and child poverty.
The five bills are Harris’s LIFT the Middle Class Act and Rent Relief Act; Brown and Khanna’s GAIN Act; Booker’s HOME Act; and Brown and Bennet’s American Family Act. And according to Columbia analysis, all five would significantly cut poverty.
All of them cost a lot — but they all cost about the same or less than the recent round of Republican tax cuts. All five are more ambitious than any cash proposals during the 2016 primary, or any other Democratic primary I can remember. And in a head-to-head matchup of all five, two plans (surprisingly) stood out as doing the most to reduce poverty at the least cost: bills from Cory Booker and Kamala Harris subsidizing rent for low-income households.
But all five are worth consideration. Let’s dive into each and how they stack up against one another.
The five bills, explained
Every one of the five proposals reduces poverty, and child poverty, substantially. But the proposals differ in cost and in the scale of their effect on poverty rates. The most effective bill reduces poverty by 1.3 percentage points more than the least effective bill, meaning it would lift some 4.2 million additional people out of poverty.
To compare the effects of each plan, the researchers set a baseline poverty rate. They projected a baseline poverty rate of 13.5 percent, a deep poverty rate of 4.9 percent, and a child poverty rate of 14.8 percent, before any programs are enacted. (The researchers use the supplemental poverty measure — a more accurate metric the Census Bureau releases as an alternative to the official poverty measure — as their bar.)
They then estimated how much each of the five proposals would change these baseline poverty numbers. Here’s what they found:
The differences in effects can be explained by the fact that the bills all attempt to do different things. Two bills are effectively massive expansions of the EITC, which provides supplemental cash to low-income people who work:
- The LIFT (Livable Incomes for Families Today) the Middle Class Act ($247 billion a year) by Kamala Harris would add another credit tied to work paid out monthly at a rate of $3,000 a year for childless single adults and $6,000 a year for couples or parents, phasing out for childless singles who make at least $50,000 a year and couples and parents who make $100,00.
- The GAIN (Grow American Incomes Now) Act ($111 billion a year) by Sherrod Brown and Ro Khanna would double or triple the EITC’s benefits, depending on family size, and increase the maximum income for families to receive the credit from the mid-$40,000 range to as high as $75,000.
Two bills are meant to provide rental assistance by giving away cash. (Elizabeth Warren also has an affordable housing bill, but it spends money directly on building low-cost housing, rather than distributing it to renters as cash the way the below bills do, so we left it out of this analysis.):
- The Rent Relief Act ($93 billion a year), also by Harris, would offer a refundable tax credit to people making $100,000 or less and spending at least 30 percent of their income on rent. The credit would be worth a certain percentage of the difference between their rent (capped at 150 percent of area fair market rent) and 30 percent of their income. For the poorest renters, the credit would cover the full difference; for slightly less poor renters, 75 percent, and so on.
- The HOME (Housing, Opportunity, Mobility, and Equity) Act ($134 billion a year) by Booker would, like the Harris bill, provide a refundable credit to people paying more than 30 percent of their income in rent. The credit would be worth the difference between their year’s rent (capped at the area fair-market rent) and 30 percent of their income. Unlike the Harris bill, there’s no strict income requirement, though the fair market rent requirement prevents the credit from going to luxury renters.
Finally, one bill would establish a child allowance, a policy common in rich countries wherein parents receive a regular cash stipend for each child they’re raising, both to reduce poverty and to defray the costs of raising the next generation:
- The American Family Act ($91 billion a year), by Sherrod Brown and fellow potential candidate Michael Bennet, would provide $3,000 a year for kids 6 to 18, and $3,600 per year for kids 0 to 5, paid out monthly to everyone earning under $75,000 (single parents) or $110,000 (couples), after which the credit would phase out. The version of the bill set to be introduced in Congress will feature higher phaseout thresholds, the Columbia team (which is in touch with the staff drafting it) says, so the Columbia researchers modeled it with phaseouts at $130,000 for single parents and $180,000 for couples.
How the bills stack up
Every one of these five bills would lift millions of Americans out of poverty. Even the “least” effective one — the GAIN Act — would reduce the poverty rate by 2 points, which means lifting 6.6 million people out of poverty.
But each one approaches that task differently. For instance, the American Family Act by Brown and Bennet, which pays out $250 to $300 a month per child to low- and middle-income parents, would cut child poverty especially deeply. The American Family Act also cuts deep poverty more than the other plans, suggesting that deep poverty is concentrated among families with kids.
The bill that makes the biggest dent in overall poverty was Harris’s LIFT Act. That makes sense, as it’s by far the most expensive. But the second most effective — and this was a surprise to me — is Cory Booker’s HOME Act, a rental subsidy bill that provides tax credits to rent-burdened households. (More on this below.)
The researcher team at Columbia — Christopher Wimer, Sophie Collyer, Robert Paul Hartley, and Sara Kimberlin — also wanted to find out how the bills compared on a cost-equivalent basis.
That is, Bennet-Brown’s child tax credit plan costs less than half what Kamala Harris’s LIFT Act would cost every year. If they were both scaled to cost the same, what would be the impact?
So the Columbia team produced cost-equivalent estimates of the bills’ effects on poverty, estimating what the bills would do to poverty if they all cost the same amount. In order to keep the comparisons simple, they keep the proposed beneficiaries constant and either inflate or deflate the net benefits. Here’s how the plans stack up if they all cost $247 billion a year, the cost of the most expensive proposal (Harris’s LIFT Act):
Here, too, Booker’s HOME Act fares extremely well — but even it is bested by Kamala Harris’s other cash proposal: the Rent Relief Act, which, like the HOME Act, would give tax credits to households paying high rents. Giving money in the form of rental assistance appears to be more efficient, dollar for dollar, in reducing poverty than handing it out with more strings attached.
Who benefits from these plans?
Another way to weigh the bills against each other is to compare how much of their cost goes to each segment of the US income distribution: those in deep poverty (less than half the poverty line); those in poverty but not deep poverty (50 to 100 percent of the poverty line); those near poverty (between 100 and 150 percent of the poverty line); those who are not poor but still low-income (between 150 percent and 200 percent of the poverty line); and lower-middle-, middle-, and high-income people making twice the poverty line or above.
Everyone low-income and below is struggling, and the “higher income” category still includes many lower-middle and middle-class people who could use assistance. But mapping out where the benefits go among these groups gives you a sense of how targeted each plan is to the ultra-poor versus the working class generally.
This analysis reveals that the rental assistance bills are very well-targeted to the poor and ultra-poor:
Harris’s Rent Relief Act costs less than half as much as her LIFT Act. But while the latter provides $39.1 billion a year to people in poverty, the former provides $42 billion. It’s vastly cheaper, and yet it directs more money to people in poverty.
Booker’s HOME Act, which costs significantly more per year than Harris’s rental bill, accordingly provides $51.9 billion to households in poverty, the most of any plan considered.
Going into this analysis, I honestly expected the Bennet-Brown child benefit bill to be the most focused for reducing poverty. It has no phase-in or work requirement the way the LIFT and GAIN Acts do, so it’s available to the poorest parents in a way that existing tax credits are not at all.
But the rental bills outdid it. Columbia’s Robert Paul Hartley notes, “The rental plans’ effects on poverty make a lot of sense when you think about how poor renters are relative to owners, and how GAIN and LIFT are really geared toward the middle class” and making EITC benefits available to the middle class. Fellow researcher Sara Kimberlin adds that housing and utilities “generally make up the largest share of total basic needs expenses for most families.” A targeted credit meant to address those expenses would, thus, be expected to have a very strong effect on poverty.
To some extent, the problem being identified here is the same one that sociologist Matthew Desmond explicated in his excellent book Evicted. Desmond notes that even in relatively low-rent cities like Milwaukee (as opposed to hyper-expensive coastal cities like San Francisco or New York), the constant burden of rent and the additional burdens and costs of being evicted for failure to pay helps trap poor families and prevent them from escaping both poverty itself and poor, low-opportunity neighborhoods. (Desmond endorsed the Harris rental credit bill in a statement released upon the bill’s unveiling.)
The rental subsidy bills, however, could have an unintended consequence that hampers their poverty-reducing potential: They could raise rents, particularly in cities like New York or San Francisco, where demand for housing vastly outstrips supply. Rental subsidies, by definition, increase demand for housing. That doesn’t have to lead to an increase in rents if developers respond by increasing housing supply in turn, but in many American cities and particularly suburbs, bans on multi-family housing (a.k.a. “single-family zoning”), onerous parking requirements, height limitations, and other zoning measures combine to artificially strangle the construction of new apartment units.
In that environment, a sudden surge in demand for housing due to rental credits could lead to increased market rents all around. The Harris and Booker bills would adapt to this reality, because the size of their credits is pegged to prevailing rents in the area where the beneficiary lives. But if the credit grows in response to a spike in rents that it generated, that increases the price tag of the program from the federal government’s perspective and risks setting off a vicious cycle. Booker’s bill includes inducements to local governments to adopt laxer zoning rules, which could mitigate this problem, but Harris’s bill does not include that component.
Hartley, at Columbia, argued in an email to me that rental costs are such a huge part of low-income budgets that policies increasing the EITC or child tax credit could have the same problem: They’d boost demand for housing as well. He also noted that Booker’s and Harris’s bills distribute the rental credit on a yearly basis. If people see it as part of their tax return, that might disincline recipients to use it to find and pay for a higher-rent apartment.
He also notes that the most recent empirical study on the effect of rent subsidies in the US on rental prices — by economists Michael Eriksen and Amanda Ross — found small effects even in supply-constrained markets, though as Brookings Institution economist Jenny Schuetz added on Twitter, we’ve never had subsidies on the scale of the Booker or Harris proposals before.
These bills are meant to do more than reduce poverty
This analysis does not tell you everything you need to know to evaluate these proposals. These plans have goals other than poverty reduction.
Harris’s LIFT Act and Brown and Khanna’s GAIN Act, for instance, are meant to make work more lucrative for low- and middle-income families who aren’t necessarily poor. The Bennet-Brown child benefit is meant to go to even upper-middle-class families and defray child care expenses. The rental assistance bills are, first and foremost, housing policy.
There are also other bills that will likely have an impact on poverty that the Columbia team has not yet modeled. Primary among these are a job guarantee proposal from Bernie Sanders and an actual bill from Cory Booker (backed by New York Sen. Kirsten Gillibrand), and Elizabeth Warren’s housing bill, which targets affordable housing construction rather than sending money directly to renters. The candidates are also likely to run on big spending plans that focus their benefits on the middle class and probably won’t make a big dent in poverty, like free college or paid family leave. Medicare-for-all would likely reduce poverty somewhat, but that depends a lot on how it’s financed, and in particular on whether poor people now getting Medicaid have to pay new taxes.
It’s also important to keep in mind that many candidates have endorsed each other’s proposals. Gillibrand, Warren, Sen. Jeff Merkley (D-OR), and Kamala Harris have endorsed Booker’s job guarantee bill; Gillibrand and Merkley have endorsed Harris’s Rent Relief Act; Gillibrand, Warren, Booker, Harris, and Sen. Amy Klobuchar (D-MN) have all endorsed the Bennet-Brown child benefit bill. These aren’t either/or propositions; it is possible to enact multiple of these plans. Indeed, Harris and Sherrod Brown have each sponsored two.
That means the 2020 primaries could see a virtuous cycle as candidates compete to see who can put together the more effective anti-poverty platform, with the most resources going to the deeply poor. The competition is already off to a good start.
Sign up for the Future Perfect newsletter. Twice a week, you’ll get a roundup of ideas and solutions for tackling our biggest challenges: improving public health, decreasing human and animal suffering, easing catastrophic risks, and — to put it simply — getting better at doing good.